Straight answers for your security review.
No badge theater. This page states exactly how Recon handles data today, so your RevOps and IT teams can evaluate it honestly.
The posture, in plain terms.
Public-source intelligence only
Recon researches from verified public sources — filings, registries, news, patents, hiring data. It does not scrape private systems or buy gray-market contact data. Every claim carries its source.
API keys stay server-side
All model and data-source API keys live on the backend. Nothing sensitive ships in the desktop client. LLM calls are made server-side only.
Your identity is yours
Your company config — practices, case studies, competitors, proof points — is tenant-scoped and used only to personalize your outputs. It is never shared across customers.
No training on your data
Your configuration and research outputs are not used to train models. Model providers are used with training-disabled API settings.
Full attribution trail
Every claim in every dossier links to the source that produced it, with a trust weight and a timestamp. If a claim can't be sourced, it doesn't ship — that's a product guarantee, not a policy.
Controlled pilot deployments
Pilots run on dedicated infrastructure with per-tenant isolation, spend caps, and access controls. Deployment options — including running closer to your environment — are discussed during pilot scoping.
Certifications, honestly.
We're an early-stage product. We do not yet hold SOC 2 or ISO certifications, and we won't put badges on this page until we do. What we offer today: transparent architecture, public-source-only data, tenant isolation, and direct access to the team building it — bring your security questionnaire to the pilot conversation and we'll answer every line.
Bring your security questionnaire.
We'd rather answer hard questions early than surprise you late.
Book a Pilot